Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.stecf.org/conferences/adass/adassVII/reprints/gaffneyn.ps.gz Äàòà èçìåíåíèÿ: Mon Jun 12 18:51:45 2006 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 03:47:42 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï

Astronomical Data Analysis Software and Systems VII
ASP Conference Series, Vol. 145, 1998
R. Albrecht, R. N. Hook and H. A. Bushouse, e
Ö Copyright 1998 Astronomical Society of the Pacific. All rights reserved.
ds.
The Phase II Language for the Hobby # Eberly Telescope
Niall I. Ga#ney
Hobby # Eberly Telescope, RLM 15.308 University of Texas, Austin, TX
78712, Email: niall@astro.as.utexas.edu
Mark E. Cornell
McDonald Observatory, RLM 15.308 University of Texas, Austin, TX
78712, Email: cornell@puck.as.utexas.edu
Abstract. We describe our recent software e#ort to enable queued ob­
servations on the Hobby # Eberly Telescope. For the past year, we have
worked on development and implementation of a ``Phase II'' style obser­
vation submission and execution system. We have developed a language
in which investigators can describe their observations, a simple method
for submission, and tools to execute observations. We discuss the special
features of this language that were implemented to accommodate some
of the unique characteristics of the Hobby # Eberly Telescope.
1. Introduction
The Hobby # Eberly Telescope (HET) (Sebring & Ramsey 1994) is a 9.2 meter
optical telescope currently undergoing scientific commissioning in west Texas.
The telescope is operated by McDonald Observatory on behalf of the HET
consortium 1 . Once fully operational, this telescope will operate in a queue sched­
uled, service observing mode 85% of the time.
What distinguishes the HET from other optical telescopes is its fixed eleva­
tion of 55 # . The telescope acquires and tracks an object by rotating in azimuth
and moving an Arecibo­style tracker mechanism across the focal surface. While
this procedure allows the HET to access over 70% of the sky over the course of
the year, each object is only observable for approximately 1 hour once or twice
a night. We therefore maximize the productivity of this telescope with queue
scheduled service observing. To facilitate queue scheduling, we have constructed
a scripting language in which investigators can describe their observations in fa­
miliar astronomical terms. Using these observing scripts, the HET Operations
Sta# can build a queue of observations and then execute the observations as
required by the investigator.
1 The HET consortium members are The University of Texas at Austin, The Pennsylvania
State University, Stanford University, Ludwig--Maximilians--Universit˜at M˜unchen, and Georg--
August--Universit˜at G˜ottingen.
284

The Phase II Language for the Hobby # Eberly Telescope 285
2. The Phase II Language
Because of the wide variety of computers in use by our investigators, we needed
to develop a platform--independent Phase II system. We chose to use a text­
based system of keywords and parameters for the language. These keywords
are contained within modular ``templates'', each of which represent a simple
observational concept. These templates are:
. Summary template -- Investigators' contact information and projects' sci­
entific goals
. Object/Calibrator template -- Information about the object (e.g., position,
magnitude)
. Instrument template -- Setup information for the instrument
. Constraints template -- Conditions acceptable for observations (e.g., sky
brightness)
. Lamp/Flat/Dark template-- Setup information for unique calibrations re­
quired for the data
. Plan template-- The sequence of actions to be taken to acquire data
The investigator is able to specify details such as which dates and times
are acceptable for the observations, what observing conditions are acceptable,
how periodic or phase critical observations are to be carried out, and what
the relative priority of di#erent parts of the project are to the overall scientific
goal. All of this information is specified in terms familiar to an astronomer with
concepts similar to what one would tell an experienced observer unfamiliar with
the specifics of the project.
Each template can be called many times from within one plan or from
multiple plans, saving the investigator from having to make multiple entries
for the same object or instrument setup. In addition to this reusability, we
have implemented macros to reduce repetitive typing. Recursive macro calls
also allow the user to create complex plans with minimal typing. Details of the
language are further described in the Phase II manual 2 .
3. Unique Features
Because of the constrained motion of the HET, we needed to develop two unique
features for our Phase II language. The first of these is how one requests an
observation to be done. One can do an observation, which means that the
observations are to be executed in the given sequence with no other observations
intervening. Alternately one can schedule an observation. This allows the HET
queue to place that observation at any time in the night that is convenient
and allows other observations (possibly from other investigators) to take place
2 http://rhea.as.utexas.edu/HET documentation

286 Ga#ney and Cornell
begin object: SAO10 begin plan: SAO10
description: protogalaxy instrument: lrs setup1
ra: 00:04:45 constraint: very good
dec: +22:43:26 class: required
equinox: J2000 # starts on the 18th of November
pa: 142.3 # ±3.5 days
ra o#set: 10.3 utdate start: 18/11/1997 delta 3 12:00
dec o#set: ­23.4 do­flat: 50 5
# magnitude of acquisition object do­object: SAO10 800 scaled 2
v: 12.4 do­lamp: ThAr 10 5
# flux of object schedule­dark: 500 5
flux 8300 continuum 21.4 schedule­calibrator: bluestar 50 fixed 1
rv: 430.3 # wait 3 ±1 days
finding chart: sao10.fit schedule­wait: 3 delta 1
end object: do­object: SAO10 800 scaled 2
end plan:
Table 1. An example object template (left) and plan (right). Lines
preceded by the ``#'' character are comments.
during the interim. The plan in the right column of Table 1 demonstrates both
of these commands. The sequence of ``do'' commands require that a flat lamp
be taken immediately before and a wavelength calibrator be taken immediately
after the observations of the object. The first two ``schedule'' commands allow
the HET Operation sta# to make those observations at any time in the night.
The schedule­wait command tells the observer to wait for a period of time,
allowing other observations to take place during the interim.
The second new feature allows the investigator to give either static or dy­
namic integration times. As the HET tracks an object across its 11 meter
primary mirror, the 9.2 meter beam becomes vignetted. This vignetting is a
function of the location of the object on the sky, the time the object was ac­
quired, the length of the track, and the azimuth of the telescope. Because the
investigator will typically not know when his/her observation will be scheduled,
we allow the investigator to either describe an exposure time that will scaled
according to this vignetting function and the observing conditions at the start
of the exposure, or to statically set the integration time (e.g., if the exposure
is time critical). Exposure time scaling is done to conserve the signal­to­noise
ratio of the observation based on the vignetting function and estimates of the
atmospheric transmission and seeing losses.
4. Plan Submission and Execution
Once plans are created by the investigator, they are e--mailed to the HET Op­
erations Team in Austin. Upon receipt, the plans are automatically validated
using a procmail--type 3 validation system. Approved plans are then written to
disk where, at a fixed time each day, any changes and additions are compiled into
the HET Observation relational database. This database is a Starbase database
(Roll 1996) to which we have added some new HET--related calculations.
3 http://www.ling.helsinki.fi/ reriksso/procmail/mini­faq.html

The Phase II Language for the Hobby # Eberly Telescope 287
Figure 1. The plan execution tool reads in plan files and coordinates
the observations with the other subsystems.
The database is transfered over the Internet to the HET site at the begin­
ning of each night. The database will then be used by the resident astronomer to
schedule plans throughout the night (Ga#ney & Cornell 1997). When a sched­
uled target becomes observable, the plan is loaded into the plan execution tool
(Figure 1). This tool displays the relational database information in an easy--to--
digest GUI format. This GUI sends coordinates to the telescope control system
and creates macros to be executed on the instrument control systems (ICS) to
acquire the data. When the data is written to disk by the ICS, this tool logs the
data as being acquired, appends comments to and fixes up the FITS headers to
match the standard header format for the HET, moves the data to the correct
directory, and queues the data for backup and transfer to Austin.
References
Ga#ney, N. I., & Cornell, M. E. 1997 in ASP Conf. Ser., Vol. 125, Astronomical
Data Analysis Software and Systems VI, ed. Gareth Hunt & H. E. Payne
(San Francisco: ASP), 379
Roll, J. 1996, in ASP Conf. Ser., Vol. 101, Astronomical Data Analysis Software
and Systems V, ed. George H. Jacoby & Jeannette Barnes (San Francisco:
ASP), 536
Sebring, T. A., & Ramsey, L. W. 1994, in Advanced Technology Optical Tele­
scopes V, SPIE Tech Conf. 2199